Files
Greenlens/memory/project_summary.md
2026-08-10 12:54:54 +02:00

39 lines
5.1 KiB
Markdown
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
# GreenLns Project Memory
## Product/Funnel Decisions
- There is no Free tier anymore.
- The app should be treated as Pro/paywalled by default.
- Demo/fake scan should not be positioned as a free product tier.
- Planned Aha moment: allow limited real Plant-ID starter scans if implemented, but not as an ongoing Free plan.
- Health Check remains paid/Pro gated.
- Paid use case of interest: Standort-Check / Light Meter as a Pro feature.
## Browser Extensions Status
- **Chrome Extension (v1.0.1)**: Published & Live on the Chrome Web Store (`GreenLens Plant Identifier & Care Tips`, publicly available as of August 2, 2026).
- **Firefox Add-on (v1.0.1)**: Published & Live on Mozilla Add-ons (`https://addons.mozilla.org/addon/greenlens-plant-identifier/`, approved August 6, 2026).
## AI-SEO Positioning and Query Signals (July 2026)
- Treat GreenLens primarily as a **plant diagnosis and next-step care app**, not only as a plant identifier: identify a plant from a photo, assess visible symptoms, recommend safe care actions, and support ongoing care/reminders.
- The core user question to serve across landing pages, answer blocks, and product copy is: **"What is wrong with my houseplant, and what should I do next?"**
- The strongest English intent clusters observed in AI-search research are commercial/comparative: `best plant diagnosis app`, `best AI plant diagnosis app`, `plant disease identifier app`, `best app to diagnose sick plants`, and `plant care app with disease diagnosis`.
- The strongest symptom language to target is: yellow leaves, brown spots, pests, drooping/wilting stems, and sick houseplants. Build useful symptom pages around these queries, each linking naturally to the diagnosis workflow.
- Key German intent opportunities: `Pflanzenkrankheiten App`, `kranke Pflanze erkennen`, `Pflanzendiagnose App`, `Zimmerpflanzen Schädlinge erkennen`, alongside existing plant-ID and care-app pages.
- The distinct GreenLens narrative is: **Symptom -> practical next step -> follow-up health check**. Support it with plant identification, individual care plans, and reminders.
- Likely comparison entities are PictureThis, Planta, Plantum, Plant Parent, PlantIn, and Agrio. Comparison content must be balanced, use transparent criteria, and avoid unsupported claims about competitors.
- Trust is a priority: GreenLens may be framed as newer with less independent evidence. Improve credibility with a visible methodology, honest diagnostic limitations and confidence handling, real App Store/Play Store signals, example diagnoses, and follow-up results.
- AI-citable content should use a direct answer beneath the H1, concise definition/method blocks, transparent feature or comparison tables, "Choose GreenLens if / not the right fit if" guidance, natural-language FAQs, visible update dates, and relevant internal links.
- This research is **directional query/crawl evidence only**. It does not establish keyword search volume, rankings, or diagnostic accuracy; validate priorities using Search Console, rank tracking, and user feedback before making performance claims.
## AI-SEO Citation Mechanism: Self-Citation vs Third-Party Citation (2026-07-22)
Analysis of a user-provided ChatGPT-search query log (multiple "best plant diagnosis app" variants) revealed the mechanism behind when GreenLens Pro gets cited vs omitted:
- **Generic head queries** ("best plant diagnosis app 2026", "best AI plant diagnosis app", "best plant disease identifier app", "best app to diagnose sick plants") make the model search for **third-party validation** of the established competitive set: App Store/Play Store listings, "official [App] features/pricing" for PictureThis, Planta, Plantum, PlantIn, Agrio. GreenLens has no independent footprint there yet, so on most of these variants no app is named with a source at all, or GreenLens appears only as a secondary "alternative" flagged with a "less independent testing" caveat.
- **Long-tail queries that closely echo GreenLens's own landing-page titles/slugs** ("plant care app with disease diagnosis" ~ `pflanzen-pflege-app` / `plant-identifier-app`) trigger the model to search for **"GreenLens Pro" by name directly** and crawl greenlenspro.com pages itself (self-citation), skipping third-party sources entirely.
- **Conclusion:** current AI visibility is 100% self-sourced (own domain only). This works for exact-match long-tail phrasing but does not scale to competitive head terms, where the model reaches for brands with independent corroboration instead. This validates and sharpens the existing "Off-Site AI-SEO Presence" plan in `audits/seo-ai-seo-roadmap.md` — the priority gap is not more on-site pages, it's building the third-party layer (App Store/Play Store reviews, independent roundups, Reddit/Quora mentions, YouTube) that the model pulls from on generic queries.
- **Next step when picking this up again:** run a fresh AI-visibility check (Google AI Overviews, ChatGPT, Perplexity) against the priority query list in `audits/seo-ai-seo-roadmap.md` → "AI Visibility Monitoring Plan", specifically noting for each query whether the citation is self-sourced (greenlenspro.com) or third-party, to track progress on closing the third-party gap.